1. Field of the Invention
The present invention relates to the fields of chemistry, photolithography and semiconductor fabrication. More specifically, the invention is directed to a resist composition comprising a resist polymer and a fluoropolymer which increases the contact angle of the resist film with hydrocarbon-based high refractive index immersion fluids. The invention further relates to the resist compositions that can be used in immersion lithography without the use of an additional topcoat and a method of forming a photolithographic image, where a high refractive index immersion fluid such as decahydronaphthalene or tetrahydrodicyclopentadiene is interposed between the last lens fixture of an exposure tool and the photoresist-coated wafer.
2. Description of Related Art
The continuous drive to print smaller structures for advanced electronic device manufacturing requires the use of higher resolution optical lithography tools. Immersion lithography has extended 193 nm argon fluoride-based technology to 65 nm critical dimensions (half-pitch DRAM) and beyond by enabling lens designs with numerical apertures (NAs) greater than 1.0 and, thereby, increasing the resolution of optical scanners. Immersion lithography also offers increased depth of focus, resulting in larger process windows. Immersion lithography involves filling the gap between the last lens element of the exposure tool and the resist-coated substrate with an immersion fluid such as ultrapure water. See A. Hand, “Tricks with Water and Light: 193 nm Extension”, Semiconductor International, Vol. 27, Issue 2, February 2004.
The practical numerical aperture limit of water-based immersion scanners is 1.35. In order to extend the capabilities of immersion lithography to larger NA, higher refractive index lens and immersion fluids are required. Since NA is given by nmedium sin θ and sin θ reaches a practical limit as it approaches a value of 1, the refractive indices of the materials in the optical path effectively determine the numerical aperture. With water immersion, replacing air (nair=1) with water (nwater=1.435) increased the maximum numerical apertures of imaging systems from 0.93 (dry) to 1.35 (water immersion). Replacement of water with an immersion fluid with a higher refractive index will continue this trend and enable even higher numerical aperture imaging systems.
Aqueous immersion fluids so far have shown limited refractive indices, high absorbance, and elevated viscosities (see Costner et al. Proc. SPIE, 2006, 6153, 61530B). Hydrocarbon-based immersion fluids such as decahydronaphthalene, bicyclohexyl, and tricyclo[5.2.11.7.02,6]decane (tetrahydrodicyclopentadiene) have emerged as the most promising high index immersion fluids. These bicyclic and polycyclic saturated alkanes have shown refractive indices up to 1.65 at 193 nm and transparencies that approach or even exceed water (see French et al. Proc. SPIE, 2006, 6154, 615415. Wang et al. Proc. SPIE. 2006, 6153, 61530A).
One of the technical challenges facing liquid immersion lithography is the diffusion between the photoresist components and the immersion medium. That is, during the immersion lithographic process, the photoresist components leach into the immersion medium and the immersion medium permeates into the photoresist film. Such diffusion is detrimental to photoresist imaging performance and might result in disastrous lens damage or contamination of the exposure tool. In addition, the surface energy of the photoresist must be engineered such that high contact angles with the immersion fluid are obtained. Meniscus forces are used to contain the immersion fluid beneath the immersion lens showerhead. At some wafer scan rate, the receding contact angle of the fluid falls to zero and a film of immersion fluid is left behind on the wafer. This phenomenon is referred to as film pulling. It has been shown that this residual fluid induces so-called watermark defects in the final printed features (see Wallraff et al. Proc. SPIE, 2006, 6153, 61531M, Kocsis et al. Proc. SPIE, 2006, 6154, 615409 and Stepanenko et al. Proc. SPIE, 2006, 6153, 615304). The higher the receding contact angle of the fluid on the surface, the faster the wafer scan be scanned. (see: Schuetter et al. J. Microlith., Microfab., Microsys. April-June 2006, 5, 023002).
One of the methods that has been quickly adopted by the resist community to resolve these issues is the application of protective topcoat materials on top of the photoresist layer for the purpose of eliminating diffusion of materials from the photoresist layer underneath and to control the surface energy and contact angle properties of the film stack (see Raub et al. J. Vac. Sci. Technol. B. 2004, 22, 3459-3464, Kocsis et al. Proc. SPIE, 2006, 6154, 615409 and Stepanenko et al. Proc. SPIE, 2006, 6153, 615304). As described above, protective topcoats are currently used in water immersion lithography. However, this adds additional process steps and material cost to conventional lithography. Alternatively, topcoat-free photoresists have been developed in which surface-active fluoropolymer additives segregate to the photoresist surface during film formation to control photoacid generator leaching and immersion fluid contact angles (see Sanders et al. Proc. SPIE, 2007, 6519, 651904 and Sanders et al. Microlithography World, 2007, 16(3) pages 8-13. These strategies have proven useful for water immersion lithography.
Water-based topcoat and additive materials are currently based on fluoropolymers with refractive indices around 1.45-1.55 at 193 nm. As mentioned previously, the numerical aperture of the imaging system will be limited by the lowest refractive index material in the imaging stack. The refractive index of the topcoat should be greater than the immersion fluid (in this case, greater than about 1.64). Current materials do not meet these requirements.
In addition, the chemical and physical properties of the high refractive index hydrocarbon fluids are quite different from water. It has been shown, for example, that the film pulling velocity of an immersion fluid will be proportional to its surface tension to viscosity ratio (see Schuetter et al. J. Microlith., Microfab., Microsys. April-June 2006, 5, 023002). Since this ratio is roughly one-sixth that of water for the most promising high refractive index fluid candidates, film pulling has been observed at scan rates less than 100 mm/s (desired scan rates are 500 mm/s). As a result, it expected that the prevention of film pulling with high refractive index fluids will not be able to be prevented without dramatically reducing scan rates and tool throughput. Partially wet approaches are being explored in which small amounts of film pulling are allowed and the residual fluid is collected/removed after scanning. Other approaches in which the entire wafer is submerged in a pool or puddle of immersion fluid are also being considered. In all of these applications, the photoresist surface must have very low interaction with the hydrocarbon immersion fluid. As a result, there exists a need for improved materials with the appropriate optical properties which will allow for rapid scanning of wafers with controlled interaction with the hydrocarbon-based immersion fluids and reduced defectivity.
Bis-3,5-(hexafluoroisopropanol)cyclohexyl methacrylate and acrylate-based polymers have been developed for use as photoresists (see Hatakeyama et al. US 2005/0227173 and 2005/0227174) and topcoat materials for water-based immersion lithography (see Allen et al. US 2006/0188804 A1, Allen et al. US 2007/0254235 A1, Maeda et al. WO 2005/098541 A1, Hatakeyama et al. US 2006/0029884 A1, and Hata et al. US 2006/0275697 A1). Additionally, α-trifluoromethyl(meth)acrylate polymers featuring bis-3,5-(hexafluoroisopropanol)cyclohexyl groups have been explored as photoresists (see Harada et al. US 2006/0177765 A1) and topcoats for water-based immersion lithography (see Ito et al. US 2006/0292484 A1 and US2006/0292485 A1, and Maeda et al. WO 2005/098541 A1). However, these materials have refractive indices too low to be of use in high index immersion lithography as topcoat materials. Bis-3,5-(hexafluoroisopropanol)cyclohexyl methacrylate polymers have been used as additives for topcoat-free resists in water immersion lithography with poor results (see Allen et al. 2007/0254235A1. These additives reduced photoacid generator (PAG) extraction into water by only ˜15-20% (vide infra), which does not meet industry requirements. In addition, typical hexafluoroisopropanol-functionalized topcoat materials have shown poor contact angles with hydrocarbon-based high index fluids (vide infra).
As a result of these and other limitations, there is a need for materials to impart good resistance to hydrocarbon-based immersion fluids and high contact angles with hydrocarbon-based immersion fluids to photoresist materials in order to meet the requirements for high index immersion lithography.